These studies will test the hypothesis that with oncogenic transformation the secretion of lysosomal proteases by transformed cells results from alterations in carbohydrate structure and/or placement on the glycoproteins. We plan to investigate the possible relationship between carbohydrate structure and protease secretion by the following approaches. First, Cathepsin B, Cathepsin D, and/or Beta-glucuronidase will be isolated from the medium of transformed cells and will be compared to similar fractions secreted by nontransformed, chloroquine-treated cells with respect to structure, number, and location of oligosaccharide moieties. We are interested in determining whether the protease from transformed cells lack the mannose-6-PO4 recognition marker required for lysosomal localization, and/or whether they possess carbohydrate structures related to those of secretory glycoproteins. Second, we will study the mannose-6-PO4 receptor protein, in normaland transformed cells, with respect to cellular distribution, localization, and carbohydrate-binding properties. Third, we plan to investigate the effects of alterations in oligosaccharides on the intracellular processing and transport of the lysosomal proteases. The functional significance of the alterations in carbohydrate moieties of the glycoproteins in transformed cells has not been investigated. Loss or modification of the mannose-6-PO4 recognition marker of lysosomal enzymes might result in their aberrant accumulation in the extracellular milieu where they could act upon the surfaces and connective tissue stroma of tumor cells. Therefore, it is important to determine whether neoplastic cells produce altered forms of lysosomal proteases which are not recognized by the usual cellular recognition system responsible for "routing" the enzymes to lysosomes. We have found that virus-transformed fibroblasts, in contrast to normal cells, exhibit increased membrane protease activity as measured by the degradation of fibronectin, laminin, and collagen (type IV). The enzyme responsible for the degradation of fibronectin is a metalloendoprotease based on product analyses and inhibitor studies. Presently, the enzyme(s) responsible for the degradation of laminin and collagen has not been characterized. (A)